This project will apply a newly-developed tracer kinetic method for estimation of regional brain benzodiazepine binding site ("benzodiazepine receptor";BZR) numbers to normal human subjects and to patients with anxiety disorders. It will employ positron emission tomography (PET) and the radiotracer [11C]flumazenil to image and quantify cerebral BZR in vivo. The project has three major divisions: the first is devoted to further model and method validation; the second concerns evaluation of benzodiazepine receptor occupancy following administration of pharmacologic doses of unlabeled benzodiazepines; the third will explore the possibility of altered benzodiazepine receptors in patients with anxiety disorders. The studies should yield new information on the central nervous system actions of benzodiazepines in humans, and will permit the direct testing of hypothesized psychiatric disease mechanism. Experiments are proposed to further investigate current modeling assumptions regarding the quantification of high-affinity ligand binding to the BZR. The initial experimental design will make use of test- retest receptor binding studies with distinct tracer input functions and analytic methods employed. Tracer kinetic estimation of receptor density, on the basis of the regional brain distribution volume of [11C]flumazenil following bolus intravenous injection, will be compared with the equilibrium tracer distribution achieved by intravenous infusion to establish steady-state levels in blood and brain. Subsequently, the precision of kinetic determinations of receptor availability will be investigated in subjects studied twice over several weeks. Analytic methods will be developed to quantify and display changes in BZR availability resulting from administration of pharmacologic doses of unlabeled benzodiazepines. The relationships between plasma levels of oxazepam, brain BZR occupancy, and behavioral and physiologic measure of drug action will be investigated in normal subjects following subacute administration of oxazepam. Evidence for dose-response relationships between plasma drug levels and PET BZR occupancy, power spectral changes in the EEG, and altered performance on neuropsychometric testing will be sought. Aspects of benzodiazepine receptor pharmacology will be evaluated in patients with anxiety disorders including those with generalized anxiety or with panic attacks. The possibility of primary alteration of BZR number will be examined in unmedicated patients. Subsequently, the possibility of altered agonist interaction with the BZR will be evaluated by determination of the BZR occupancy produced by pharmacologic doses of oxazepam.